--- title: 1. Challenge: dolphin instance segmentation model keywords: fastai sidebar: home_sidebar summary: "The goal of this challenge is to find all instances of dolphins in a picture and then color pixes of each dolphin with a unique color." description: "The goal of this challenge is to find all instances of dolphins in a picture and then color pixes of each dolphin with a unique color." nb_path: "notebooks/01_Dolphin_instance_segmentation_challenge.ipynb" ---
{% raw %} 
numpy       : 1.18.5
torch       : 1.7.1
torchvision : 0.8.2
PIL         : 7.2.0
{% endraw %}

Introduction and motivation

Fill in please

{% raw %}
{% endraw %}

Download data

We start by downloading and visualizing the dataset containing 200 photographs with one or more dolphins split into a training set containing 160 photographs and a validation set containing 40 photographs.

{% raw %} 
from dolphins_recognition_challenge.datasets import get_dataset, display_batches
    
data_loader, data_loader_test = get_dataset("segmentation", batch_size=3)

display_batches(data_loader, n_batches=2, width=600)
{% endraw %}

Data augmentation

In order to prevent overfitting which happens when the dataset size is too small, we perform a number of transformations to increase the size of the dataset. One transofrmation implemented in the Torch vision library is RandomHorizontalFlip and we will implemented MyColorJitter which is basically just a wrapper around torchvision.transforms.ColorJitter class. However, we cannot use this class directly without a wrapper because a transofrmation could possibly affect targets and not just the image. For example, if we were to implement RandomCrop, we would need to crop segmentation masks and readjust bounding boxes as well.

{% raw %} 
class MyColorJitter:
    def __init__(self, brightness=0.5, contrast=0.5, saturation=0.5, hue=0.5):
        self.torch_color_jitter = torchvision.transforms.ColorJitter(
            brightness=brightness, contrast=contrast, saturation=saturation, hue=hue
        )

    def __call__(self, image, target):
        image = self.torch_color_jitter(image)
        return image, target
{% endraw %}

We will make a series of transformations on an image and we will combine all those transofrmations in a single one as follows:

{% raw %} 
import transforms as T

def get_tensor_transforms(train):
    transforms = []
    # converts the image, a PIL image, into a PyTorch Tensor
    transforms.append(T.ToTensor())
    if train:
        # during training, randomly flip the training images
        # and ground-truth for data augmentation
        transforms.append(
            MyColorJitter(brightness=0.5, contrast=0.5, saturation=0.5, hue=0.5)
        )
        transforms.append(T.RandomHorizontalFlip(0.5))
        # TODO: add additional transforms: e.g. random crop
    return T.Compose(transforms)
{% endraw %} {% raw %} 
data_loader, data_loader_test = get_dataset("segmentation", batch_size=2, get_tensor_transforms=get_tensor_transforms)

display_batches(data_loader, n_batches=2, width=800)
{% endraw %}

With data augementation defined, we are ready to generate the actual datasets used for training our models.

{% raw %} 
batch_size = 4

data_loader, data_loader_test = get_dataset(
    "segmentation", get_tensor_transforms=get_tensor_transforms, batch_size=batch_size
)

display_batches(data_loader, n_batches=4, width=800)
{% endraw %}

{% include tip.html content='incorporate more transformation classes such as RandomCrop etc. (https://pytorch.org/docs/stable/torchvision/transforms.html)' %}

Model

{% raw %} 
def get_instance_segmentation_model(hidden_layer_size):
    # our dataset has two classes only - background and dolphin    
    num_classes = 2
    
    # load an instance segmentation model pre-trained on COCO
    model = torchvision.models.detection.maskrcnn_resnet50_fpn(
        pretrained=True
    )  # box_score_thresh=0.5

    # get the number of input features for the classifier
    in_features = model.roi_heads.box_predictor.cls_score.in_features
    # replace the pre-trained head with a new one
    model.roi_heads.box_predictor = FastRCNNPredictor(in_features, num_classes)

    # now get the number of input features for the mask classifier
    in_features_mask = model.roi_heads.mask_predictor.conv5_mask.in_channels

    model.roi_heads.mask_predictor = MaskRCNNPredictor(
        in_channels=in_features_mask, 
        dim_reduced=hidden_layer_size,
        num_classes=num_classes
    )

    return model
{% endraw %} {% raw %} 
device = torch.device("cuda") if torch.cuda.is_available() else torch.device("cpu")

# get the model using our helper function
model = get_instance_segmentation_model(hidden_layer_size=256)

# move model to the right device
model.to(device)

# construct an optimizer
params = [p for p in model.parameters() if p.requires_grad]
optimizer = torch.optim.SGD(params, lr=0.005, momentum=0.9, weight_decay=0.0005)

# and a learning rate scheduler which decreases the learning rate by
# 10x every 3 epochs
lr_scheduler = torch.optim.lr_scheduler.StepLR(optimizer, step_size=10, gamma=0.1)
{% endraw %} {% raw %} 
from traceback_with_variables import printing_tb

from engine import train_one_epoch, evaluate

# let's train it for 20 epochs
num_epochs = 20

print("Training...")
with printing_tb():
    for epoch in range(num_epochs):
        # train for one epoch, printing every 10 iterations
        train_one_epoch(model, optimizer, data_loader, device, epoch, print_freq=10)
        # update the learning rate
        lr_scheduler.step()
        # evaluate on the test dataset
        evaluate(model, data_loader_test, device=device)

Training...

/usr/local/lib/python3.6/dist-packages/torch/nn/functional.py:3103: UserWarning: The default behavior for interpolate/upsample with float scale_factor changed in 1.6.0 to align with other frameworks/libraries, and now uses scale_factor directly, instead of relying on the computed output size. If you wish to restore the old behavior, please set recompute_scale_factor=True. See the documentation of nn.Upsample for details. 
  warnings.warn("The default behavior for interpolate/upsample with float scale_factor changed "

Epoch: [0]  [ 0/40]  eta: 0:00:51  lr: 0.000133  loss: 3.8964 (3.8964)  loss_classifier: 0.9773 (0.9773)  loss_box_reg: 0.2797 (0.2797)  loss_mask: 2.5986 (2.5986)  loss_objectness: 0.0113 (0.0113)  loss_rpn_box_reg: 0.0295 (0.0295)  time: 1.2918  data: 0.7365  max mem: 4477
Epoch: [0]  [10/40]  eta: 0:00:16  lr: 0.001414  loss: 1.3453 (2.1033)  loss_classifier: 0.2631 (0.4807)  loss_box_reg: 0.2872 (0.2843)  loss_mask: 0.7485 (1.2939)  loss_objectness: 0.0135 (0.0278)  loss_rpn_box_reg: 0.0095 (0.0166)  time: 0.5513  data: 0.0742  max mem: 5189
Epoch: [0]  [20/40]  eta: 0:00:10  lr: 0.002695  loss: 1.0537 (1.5074)  loss_classifier: 0.2357 (0.3438)  loss_box_reg: 0.2662 (0.2664)  loss_mask: 0.4620 (0.8403)  loss_objectness: 0.0258 (0.0356)  loss_rpn_box_reg: 0.0098 (0.0215)  time: 0.4777  data: 0.0080  max mem: 5189
Epoch: [0]  [30/40]  eta: 0:00:05  lr: 0.003975  loss: 0.6962 (1.2475)  loss_classifier: 0.1226 (0.2692)  loss_box_reg: 0.2354 (0.2580)  loss_mask: 0.2537 (0.6471)  loss_objectness: 0.0258 (0.0364)  loss_rpn_box_reg: 0.0248 (0.0368)  time: 0.4791  data: 0.0082  max mem: 5189
Epoch: [0]  [39/40]  eta: 0:00:00  lr: 0.005000  loss: 0.6535 (1.1129)  loss_classifier: 0.1009 (0.2299)  loss_box_reg: 0.2537 (0.2589)  loss_mask: 0.2390 (0.5548)  loss_objectness: 0.0162 (0.0333)  loss_rpn_box_reg: 0.0138 (0.0359)  time: 0.4851  data: 0.0084  max mem: 5189
Epoch: [0] Total time: 0:00:20 (0.5036 s / it)
creating index...
index created!
Test:  [ 0/10]  eta: 0:00:10  model_time: 0.3962 (0.3962)  evaluator_time: 0.1608 (0.1608)  time: 1.0478  data: 0.4880  max mem: 5189
Test:  [ 9/10]  eta: 0:00:00  model_time: 0.3123 (0.3242)  evaluator_time: 0.1065 (0.1107)  time: 0.4971  data: 0.0542  max mem: 5189
Test: Total time: 0:00:05 (0.5032 s / it)
Averaged stats: model_time: 0.3123 (0.3242)  evaluator_time: 0.1065 (0.1107)
Accumulating evaluation results...
DONE (t=0.02s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.308
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.692
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.213
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.194
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.294
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.471
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.152
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.428
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.470
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.416
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.467
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.530
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.331
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.699
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.266
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.143
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.328
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.517
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.152
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.466
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.518
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.479
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.501
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.600
Epoch: [1]  [ 0/40]  eta: 0:00:49  lr: 0.005000  loss: 0.7875 (0.7875)  loss_classifier: 0.1042 (0.1042)  loss_box_reg: 0.2977 (0.2977)  loss_mask: 0.3597 (0.3597)  loss_objectness: 0.0119 (0.0119)  loss_rpn_box_reg: 0.0139 (0.0139)  time: 1.2469  data: 0.7677  max mem: 5189
Epoch: [1]  [10/40]  eta: 0:00:16  lr: 0.005000  loss: 0.6108 (0.5808)  loss_classifier: 0.0833 (0.0781)  loss_box_reg: 0.2088 (0.2200)  loss_mask: 0.2108 (0.2290)  loss_objectness: 0.0067 (0.0088)  loss_rpn_box_reg: 0.0117 (0.0450)  time: 0.5543  data: 0.0756  max mem: 5189
Epoch: [1]  [20/40]  eta: 0:00:10  lr: 0.005000  loss: 0.6018 (0.6000)  loss_classifier: 0.0833 (0.0915)  loss_box_reg: 0.2180 (0.2223)  loss_mask: 0.2108 (0.2391)  loss_objectness: 0.0062 (0.0103)  loss_rpn_box_reg: 0.0126 (0.0369)  time: 0.5115  data: 0.0070  max mem: 5189
Epoch: [1]  [30/40]  eta: 0:00:05  lr: 0.005000  loss: 0.5341 (0.5606)  loss_classifier: 0.0825 (0.0877)  loss_box_reg: 0.1813 (0.2030)  loss_mask: 0.1987 (0.2265)  loss_objectness: 0.0084 (0.0106)  loss_rpn_box_reg: 0.0160 (0.0328)  time: 0.5386  data: 0.0076  max mem: 5189
Epoch: [1]  [39/40]  eta: 0:00:00  lr: 0.005000  loss: 0.4563 (0.5389)  loss_classifier: 0.0737 (0.0846)  loss_box_reg: 0.1535 (0.1935)  loss_mask: 0.1899 (0.2218)  loss_objectness: 0.0063 (0.0107)  loss_rpn_box_reg: 0.0123 (0.0283)  time: 0.5404  data: 0.0078  max mem: 5189
Epoch: [1] Total time: 0:00:21 (0.5454 s / it)
creating index...
index created!
Test:  [ 0/10]  eta: 0:00:08  model_time: 0.2817 (0.2817)  evaluator_time: 0.0577 (0.0577)  time: 0.8407  data: 0.4985  max mem: 5189
Test:  [ 9/10]  eta: 0:00:00  model_time: 0.2295 (0.2343)  evaluator_time: 0.0434 (0.0458)  time: 0.3379  data: 0.0546  max mem: 5189
Test: Total time: 0:00:03 (0.3441 s / it)
Averaged stats: model_time: 0.2295 (0.2343)  evaluator_time: 0.0434 (0.0458)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.411
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.803
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.360
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.261
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.383
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.631
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.184
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.516
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.531
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.437
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.512
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.683
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.419
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.756
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.420
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.229
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.405
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.651
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.189
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.514
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.534
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.463
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.513
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.657
Epoch: [2]  [ 0/40]  eta: 0:00:47  lr: 0.005000  loss: 0.4882 (0.4882)  loss_classifier: 0.0790 (0.0790)  loss_box_reg: 0.1624 (0.1624)  loss_mask: 0.2121 (0.2121)  loss_objectness: 0.0085 (0.0085)  loss_rpn_box_reg: 0.0262 (0.0262)  time: 1.1912  data: 0.7036  max mem: 5189
Epoch: [2]  [10/40]  eta: 0:00:17  lr: 0.005000  loss: 0.5040 (0.4761)  loss_classifier: 0.0683 (0.0731)  loss_box_reg: 0.1348 (0.1604)  loss_mask: 0.2029 (0.1887)  loss_objectness: 0.0040 (0.0117)  loss_rpn_box_reg: 0.0113 (0.0423)  time: 0.5815  data: 0.0727  max mem: 5189
Epoch: [2]  [20/40]  eta: 0:00:11  lr: 0.005000  loss: 0.4084 (0.4441)  loss_classifier: 0.0575 (0.0700)  loss_box_reg: 0.1345 (0.1506)  loss_mask: 0.1856 (0.1862)  loss_objectness: 0.0044 (0.0092)  loss_rpn_box_reg: 0.0075 (0.0280)  time: 0.5348  data: 0.0085  max mem: 5189
Epoch: [2]  [30/40]  eta: 0:00:05  lr: 0.005000  loss: 0.4422 (0.4608)  loss_classifier: 0.0809 (0.0759)  loss_box_reg: 0.1508 (0.1583)  loss_mask: 0.1871 (0.1896)  loss_objectness: 0.0069 (0.0104)  loss_rpn_box_reg: 0.0108 (0.0265)  time: 0.5550  data: 0.0071  max mem: 5189
Epoch: [2]  [39/40]  eta: 0:00:00  lr: 0.005000  loss: 0.4422 (0.4476)  loss_classifier: 0.0846 (0.0742)  loss_box_reg: 0.1623 (0.1573)  loss_mask: 0.1801 (0.1841)  loss_objectness: 0.0056 (0.0091)  loss_rpn_box_reg: 0.0105 (0.0228)  time: 0.5611  data: 0.0068  max mem: 5189
Epoch: [2] Total time: 0:00:22 (0.5657 s / it)
creating index...
index created!
Test:  [ 0/10]  eta: 0:00:08  model_time: 0.2534 (0.2534)  evaluator_time: 0.0418 (0.0418)  time: 0.8188  data: 0.5209  max mem: 5189
Test:  [ 9/10]  eta: 0:00:00  model_time: 0.2125 (0.2198)  evaluator_time: 0.0315 (0.0325)  time: 0.3121  data: 0.0570  max mem: 5189
Test: Total time: 0:00:03 (0.3177 s / it)
Averaged stats: model_time: 0.2125 (0.2198)  evaluator_time: 0.0315 (0.0325)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.429
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.804
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.458
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.280
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.428
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.565
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.203
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.528
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.529
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.442
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.519
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.652
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.397
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.797
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.360
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.226
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.383
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.564
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.178
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.496
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.503
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.516
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.463
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.617
Epoch: [3]  [ 0/40]  eta: 0:00:51  lr: 0.005000  loss: 0.2576 (0.2576)  loss_classifier: 0.0448 (0.0448)  loss_box_reg: 0.0763 (0.0763)  loss_mask: 0.1327 (0.1327)  loss_objectness: 0.0014 (0.0014)  loss_rpn_box_reg: 0.0024 (0.0024)  time: 1.2926  data: 0.7692  max mem: 5189
Epoch: [3]  [10/40]  eta: 0:00:17  lr: 0.005000  loss: 0.3994 (0.3798)  loss_classifier: 0.0553 (0.0600)  loss_box_reg: 0.1143 (0.1225)  loss_mask: 0.1784 (0.1831)  loss_objectness: 0.0033 (0.0035)  loss_rpn_box_reg: 0.0077 (0.0107)  time: 0.5933  data: 0.0748  max mem: 5189
Epoch: [3]  [20/40]  eta: 0:00:11  lr: 0.005000  loss: 0.4097 (0.4063)  loss_classifier: 0.0623 (0.0629)  loss_box_reg: 0.1350 (0.1319)  loss_mask: 0.1610 (0.1762)  loss_objectness: 0.0035 (0.0042)  loss_rpn_box_reg: 0.0097 (0.0312)  time: 0.5446  data: 0.0064  max mem: 5189
Epoch: [3]  [30/40]  eta: 0:00:05  lr: 0.005000  loss: 0.4097 (0.4061)  loss_classifier: 0.0673 (0.0658)  loss_box_reg: 0.1386 (0.1350)  loss_mask: 0.1526 (0.1738)  loss_objectness: 0.0045 (0.0049)  loss_rpn_box_reg: 0.0132 (0.0265)  time: 0.5699  data: 0.0072  max mem: 5189
Epoch: [3]  [39/40]  eta: 0:00:00  lr: 0.005000  loss: 0.3495 (0.3914)  loss_classifier: 0.0637 (0.0642)  loss_box_reg: 0.1208 (0.1306)  loss_mask: 0.1485 (0.1694)  loss_objectness: 0.0042 (0.0046)  loss_rpn_box_reg: 0.0083 (0.0227)  time: 0.5727  data: 0.0070  max mem: 5189
Epoch: [3] Total time: 0:00:23 (0.5783 s / it)
creating index...
index created!
Test:  [ 0/10]  eta: 0:00:08  model_time: 0.2548 (0.2548)  evaluator_time: 0.0451 (0.0451)  time: 0.8060  data: 0.5032  max mem: 5189
Test:  [ 9/10]  eta: 0:00:00  model_time: 0.2177 (0.2221)  evaluator_time: 0.0369 (0.0376)  time: 0.3174  data: 0.0549  max mem: 5189
Test: Total time: 0:00:03 (0.3230 s / it)
Averaged stats: model_time: 0.2177 (0.2221)  evaluator_time: 0.0369 (0.0376)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.470
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.845
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.469
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.314
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.476
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.626
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.228
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.562
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.567
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.458
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.564
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.696
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.452
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.814
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.433
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.207
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.457
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.604
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.222
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.534
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.542
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.463
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.531
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.639
Epoch: [4]  [ 0/40]  eta: 0:00:49  lr: 0.005000  loss: 0.2011 (0.2011)  loss_classifier: 0.0350 (0.0350)  loss_box_reg: 0.0710 (0.0710)  loss_mask: 0.0920 (0.0920)  loss_objectness: 0.0011 (0.0011)  loss_rpn_box_reg: 0.0020 (0.0020)  time: 1.2265  data: 0.7473  max mem: 5189
Epoch: [4]  [10/40]  eta: 0:00:17  lr: 0.005000  loss: 0.3274 (0.3180)  loss_classifier: 0.0487 (0.0498)  loss_box_reg: 0.1070 (0.1064)  loss_mask: 0.1502 (0.1500)  loss_objectness: 0.0013 (0.0025)  loss_rpn_box_reg: 0.0065 (0.0092)  time: 0.5985  data: 0.0733  max mem: 5189
Epoch: [4]  [20/40]  eta: 0:00:11  lr: 0.005000  loss: 0.3445 (0.3306)  loss_classifier: 0.0537 (0.0540)  loss_box_reg: 0.1130 (0.1119)  loss_mask: 0.1531 (0.1532)  loss_objectness: 0.0018 (0.0027)  loss_rpn_box_reg: 0.0067 (0.0088)  time: 0.5579  data: 0.0068  max mem: 5189
Epoch: [4]  [30/40]  eta: 0:00:05  lr: 0.005000  loss: 0.3464 (0.3420)  loss_classifier: 0.0531 (0.0527)  loss_box_reg: 0.1100 (0.1104)  loss_mask: 0.1475 (0.1528)  loss_objectness: 0.0020 (0.0029)  loss_rpn_box_reg: 0.0087 (0.0232)  time: 0.5791  data: 0.0080  max mem: 5189
Epoch: [4]  [39/40]  eta: 0:00:00  lr: 0.005000  loss: 0.3464 (0.3525)  loss_classifier: 0.0532 (0.0554)  loss_box_reg: 0.1089 (0.1143)  loss_mask: 0.1521 (0.1550)  loss_objectness: 0.0038 (0.0042)  loss_rpn_box_reg: 0.0119 (0.0236)  time: 0.5813  data: 0.0081  max mem: 5189
Epoch: [4] Total time: 0:00:23 (0.5878 s / it)
creating index...
index created!
Test:  [ 0/10]  eta: 0:00:08  model_time: 0.2599 (0.2599)  evaluator_time: 0.0443 (0.0443)  time: 0.8043  data: 0.4971  max mem: 5189
Test:  [ 9/10]  eta: 0:00:00  model_time: 0.2221 (0.2264)  evaluator_time: 0.0369 (0.0378)  time: 0.3219  data: 0.0545  max mem: 5189
Test: Total time: 0:00:03 (0.3277 s / it)
Averaged stats: model_time: 0.2221 (0.2264)  evaluator_time: 0.0369 (0.0378)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.498
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.861
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.533
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.270
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.500
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.685
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.238
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.583
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.595
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.426
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.605
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.717
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.451
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.826
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.452
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.217
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.450
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.648
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.215
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.541
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.556
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.458
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.537
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.691
Epoch: [5]  [ 0/40]  eta: 0:00:47  lr: 0.005000  loss: 0.2295 (0.2295)  loss_classifier: 0.0628 (0.0628)  loss_box_reg: 0.0571 (0.0571)  loss_mask: 0.1022 (0.1022)  loss_objectness: 0.0047 (0.0047)  loss_rpn_box_reg: 0.0027 (0.0027)  time: 1.1821  data: 0.6976  max mem: 5189
Epoch: [5]  [10/40]  eta: 0:00:18  lr: 0.005000  loss: 0.2750 (0.3354)  loss_classifier: 0.0467 (0.0482)  loss_box_reg: 0.0926 (0.0966)  loss_mask: 0.1430 (0.1420)  loss_objectness: 0.0043 (0.0049)  loss_rpn_box_reg: 0.0047 (0.0437)  time: 0.6068  data: 0.0688  max mem: 5189
Epoch: [5]  [20/40]  eta: 0:00:11  lr: 0.005000  loss: 0.3444 (0.3535)  loss_classifier: 0.0517 (0.0531)  loss_box_reg: 0.1003 (0.1072)  loss_mask: 0.1504 (0.1532)  loss_objectness: 0.0043 (0.0049)  loss_rpn_box_reg: 0.0054 (0.0351)  time: 0.5674  data: 0.0065  max mem: 5189
Epoch: [5]  [30/40]  eta: 0:00:05  lr: 0.005000  loss: 0.3459 (0.3433)  loss_classifier: 0.0538 (0.0516)  loss_box_reg: 0.1261 (0.1073)  loss_mask: 0.1531 (0.1527)  loss_objectness: 0.0022 (0.0042)  loss_rpn_box_reg: 0.0093 (0.0274)  time: 0.5876  data: 0.0072  max mem: 5189
Epoch: [5]  [39/40]  eta: 0:00:00  lr: 0.005000  loss: 0.3459 (0.3465)  loss_classifier: 0.0538 (0.0529)  loss_box_reg: 0.1120 (0.1086)  loss_mask: 0.1593 (0.1541)  loss_objectness: 0.0022 (0.0047)  loss_rpn_box_reg: 0.0081 (0.0262)  time: 0.5885  data: 0.0072  max mem: 5189
Epoch: [5] Total time: 0:00:23 (0.5946 s / it)
creating index...
index created!
Test:  [ 0/10]  eta: 0:00:08  model_time: 0.2557 (0.2557)  evaluator_time: 0.0388 (0.0388)  time: 0.8047  data: 0.5073  max mem: 5189
Test:  [ 9/10]  eta: 0:00:00  model_time: 0.2154 (0.2201)  evaluator_time: 0.0313 (0.0318)  time: 0.3097  data: 0.0550  max mem: 5189
Test: Total time: 0:00:03 (0.3155 s / it)
Averaged stats: model_time: 0.2154 (0.2201)  evaluator_time: 0.0313 (0.0318)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.507
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.867
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.523
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.245
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.505
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.703
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.237
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.596
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.600
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.463
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.593
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.748
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.461
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.827
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.450
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.196
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.461
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.630
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.209
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.557
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.565
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.505
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.552
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.648
Epoch: [6]  [ 0/40]  eta: 0:00:49  lr: 0.005000  loss: 0.2911 (0.2911)  loss_classifier: 0.0398 (0.0398)  loss_box_reg: 0.1061 (0.1061)  loss_mask: 0.1406 (0.1406)  loss_objectness: 0.0008 (0.0008)  loss_rpn_box_reg: 0.0038 (0.0038)  time: 1.2280  data: 0.7378  max mem: 5189
Epoch: [6]  [10/40]  eta: 0:00:18  lr: 0.005000  loss: 0.3162 (0.3111)  loss_classifier: 0.0418 (0.0453)  loss_box_reg: 0.1036 (0.1030)  loss_mask: 0.1478 (0.1490)  loss_objectness: 0.0030 (0.0035)  loss_rpn_box_reg: 0.0087 (0.0104)  time: 0.6205  data: 0.0742  max mem: 5189
Epoch: [6]  [20/40]  eta: 0:00:12  lr: 0.005000  loss: 0.3162 (0.3023)  loss_classifier: 0.0405 (0.0431)  loss_box_reg: 0.0959 (0.0984)  loss_mask: 0.1447 (0.1466)  loss_objectness: 0.0018 (0.0026)  loss_rpn_box_reg: 0.0067 (0.0116)  time: 0.5761  data: 0.0077  max mem: 5189
Epoch: [6]  [30/40]  eta: 0:00:06  lr: 0.005000  loss: 0.3261 (0.3177)  loss_classifier: 0.0402 (0.0464)  loss_box_reg: 0.0932 (0.1000)  loss_mask: 0.1418 (0.1457)  loss_objectness: 0.0017 (0.0050)  loss_rpn_box_reg: 0.0092 (0.0205)  time: 0.5942  data: 0.0074  max mem: 5189
Epoch: [6]  [39/40]  eta: 0:00:00  lr: 0.005000  loss: 0.3042 (0.3130)  loss_classifier: 0.0426 (0.0456)  loss_box_reg: 0.0963 (0.1011)  loss_mask: 0.1211 (0.1439)  loss_objectness: 0.0029 (0.0048)  loss_rpn_box_reg: 0.0065 (0.0176)  time: 0.5955  data: 0.0070  max mem: 5189
Epoch: [6] Total time: 0:00:24 (0.6034 s / it)
creating index...
index created!
Test:  [ 0/10]  eta: 0:00:07  model_time: 0.2533 (0.2533)  evaluator_time: 0.0318 (0.0318)  time: 0.7994  data: 0.5112  max mem: 5189
Test:  [ 9/10]  eta: 0:00:00  model_time: 0.2063 (0.2121)  evaluator_time: 0.0219 (0.0236)  time: 0.2945  data: 0.0560  max mem: 5189
Test: Total time: 0:00:03 (0.3013 s / it)
Averaged stats: model_time: 0.2063 (0.2121)  evaluator_time: 0.0219 (0.0236)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.472
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.831
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.478
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.258
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.484
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.641
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.227
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.559
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.559
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.400
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.560
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.713
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.442
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.811
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.425
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.209
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.451
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.591
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.211
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.542
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.544
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.458
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.535
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.639
Epoch: [7]  [ 0/40]  eta: 0:00:49  lr: 0.005000  loss: 0.3007 (0.3007)  loss_classifier: 0.0478 (0.0478)  loss_box_reg: 0.1082 (0.1082)  loss_mask: 0.1297 (0.1297)  loss_objectness: 0.0041 (0.0041)  loss_rpn_box_reg: 0.0109 (0.0109)  time: 1.2296  data: 0.7325  max mem: 5189
Epoch: [7]  [10/40]  eta: 0:00:18  lr: 0.005000  loss: 0.2718 (0.2818)  loss_classifier: 0.0468 (0.0424)  loss_box_reg: 0.0843 (0.0876)  loss_mask: 0.1297 (0.1391)  loss_objectness: 0.0029 (0.0040)  loss_rpn_box_reg: 0.0079 (0.0087)  time: 0.6246  data: 0.0727  max mem: 5189
Epoch: [7]  [20/40]  eta: 0:00:12  lr: 0.005000  loss: 0.2933 (0.2964)  loss_classifier: 0.0451 (0.0424)  loss_box_reg: 0.0871 (0.0939)  loss_mask: 0.1357 (0.1440)  loss_objectness: 0.0020 (0.0037)  loss_rpn_box_reg: 0.0079 (0.0123)  time: 0.5807  data: 0.0073  max mem: 5190
Epoch: [7]  [30/40]  eta: 0:00:06  lr: 0.005000  loss: 0.3123 (0.3058)  loss_classifier: 0.0451 (0.0444)  loss_box_reg: 0.0973 (0.0996)  loss_mask: 0.1452 (0.1466)  loss_objectness: 0.0024 (0.0033)  loss_rpn_box_reg: 0.0090 (0.0120)  time: 0.6012  data: 0.0077  max mem: 5190
Epoch: [7]  [39/40]  eta: 0:00:00  lr: 0.005000  loss: 0.3041 (0.2994)  loss_classifier: 0.0399 (0.0427)  loss_box_reg: 0.0927 (0.0956)  loss_mask: 0.1427 (0.1417)  loss_objectness: 0.0016 (0.0031)  loss_rpn_box_reg: 0.0042 (0.0163)  time: 0.6019  data: 0.0077  max mem: 5190
Epoch: [7] Total time: 0:00:24 (0.6091 s / it)
creating index...
index created!
Test:  [ 0/10]  eta: 0:00:07  model_time: 0.2377 (0.2377)  evaluator_time: 0.0292 (0.0292)  time: 0.7844  data: 0.5143  max mem: 5190
Test:  [ 9/10]  eta: 0:00:00  model_time: 0.2071 (0.2105)  evaluator_time: 0.0210 (0.0221)  time: 0.2916  data: 0.0561  max mem: 5190
Test: Total time: 0:00:02 (0.2976 s / it)
Averaged stats: model_time: 0.2071 (0.2105)  evaluator_time: 0.0210 (0.0221)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.485
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.867
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.465
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.284
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.490
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.650
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.228
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.570
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.573
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.400
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.580
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.704
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.454
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.835
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.460
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.215
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.453
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.647
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.215
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.552
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.553
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.447
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.544
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.665
Epoch: [8]  [ 0/40]  eta: 0:00:48  lr: 0.005000  loss: 0.2784 (0.2784)  loss_classifier: 0.0362 (0.0362)  loss_box_reg: 0.0856 (0.0856)  loss_mask: 0.1511 (0.1511)  loss_objectness: 0.0010 (0.0010)  loss_rpn_box_reg: 0.0046 (0.0046)  time: 1.2250  data: 0.7315  max mem: 5190
Epoch: [8]  [10/40]  eta: 0:00:18  lr: 0.005000  loss: 0.2815 (0.2693)  loss_classifier: 0.0371 (0.0370)  loss_box_reg: 0.0856 (0.0884)  loss_mask: 0.1482 (0.1330)  loss_objectness: 0.0015 (0.0021)  loss_rpn_box_reg: 0.0054 (0.0088)  time: 0.6326  data: 0.0733  max mem: 5190
Epoch: [8]  [20/40]  eta: 0:00:12  lr: 0.005000  loss: 0.2772 (0.2753)  loss_classifier: 0.0384 (0.0394)  loss_box_reg: 0.0864 (0.0893)  loss_mask: 0.1252 (0.1322)  loss_objectness: 0.0018 (0.0032)  loss_rpn_box_reg: 0.0052 (0.0113)  time: 0.5873  data: 0.0077  max mem: 5190
Epoch: [8]  [30/40]  eta: 0:00:06  lr: 0.005000  loss: 0.2750 (0.2873)  loss_classifier: 0.0425 (0.0402)  loss_box_reg: 0.0868 (0.0913)  loss_mask: 0.1354 (0.1376)  loss_objectness: 0.0020 (0.0028)  loss_rpn_box_reg: 0.0052 (0.0153)  time: 0.6024  data: 0.0078  max mem: 5190
Epoch: [8]  [39/40]  eta: 0:00:00  lr: 0.005000  loss: 0.2874 (0.2828)  loss_classifier: 0.0406 (0.0399)  loss_box_reg: 0.0969 (0.0902)  loss_mask: 0.1381 (0.1350)  loss_objectness: 0.0014 (0.0029)  loss_rpn_box_reg: 0.0064 (0.0148)  time: 0.6029  data: 0.0078  max mem: 5190
Epoch: [8] Total time: 0:00:24 (0.6124 s / it)
creating index...
index created!
Test:  [ 0/10]  eta: 0:00:07  model_time: 0.2368 (0.2368)  evaluator_time: 0.0268 (0.0268)  time: 0.7932  data: 0.5267  max mem: 5190
Test:  [ 9/10]  eta: 0:00:00  model_time: 0.2076 (0.2102)  evaluator_time: 0.0200 (0.0213)  time: 0.2919  data: 0.0577  max mem: 5190
Test: Total time: 0:00:02 (0.2977 s / it)
Averaged stats: model_time: 0.2076 (0.2102)  evaluator_time: 0.0200 (0.0213)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.498
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.851
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.540
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.290
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.506
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.665
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.234
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.590
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.590
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.405
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.596
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.730
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.451
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.827
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.442
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.192
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.468
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.590
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.214
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.543
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.543
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.458
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.533
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.635
Epoch: [9]  [ 0/40]  eta: 0:00:49  lr: 0.005000  loss: 0.3162 (0.3162)  loss_classifier: 0.0453 (0.0453)  loss_box_reg: 0.1101 (0.1101)  loss_mask: 0.1526 (0.1526)  loss_objectness: 0.0014 (0.0014)  loss_rpn_box_reg: 0.0068 (0.0068)  time: 1.2318  data: 0.7385  max mem: 5190
Epoch: [9]  [10/40]  eta: 0:00:18  lr: 0.005000  loss: 0.2456 (0.2637)  loss_classifier: 0.0376 (0.0348)  loss_box_reg: 0.0793 (0.0794)  loss_mask: 0.1217 (0.1348)  loss_objectness: 0.0016 (0.0020)  loss_rpn_box_reg: 0.0034 (0.0127)  time: 0.6246  data: 0.0730  max mem: 5190
Epoch: [9]  [20/40]  eta: 0:00:12  lr: 0.005000  loss: 0.2418 (0.2679)  loss_classifier: 0.0351 (0.0346)  loss_box_reg: 0.0713 (0.0750)  loss_mask: 0.1196 (0.1299)  loss_objectness: 0.0017 (0.0020)  loss_rpn_box_reg: 0.0049 (0.0264)  time: 0.5792  data: 0.0071  max mem: 5190
Epoch: [9]  [30/40]  eta: 0:00:06  lr: 0.005000  loss: 0.2936 (0.2858)  loss_classifier: 0.0425 (0.0390)  loss_box_reg: 0.0975 (0.0865)  loss_mask: 0.1332 (0.1365)  loss_objectness: 0.0018 (0.0020)  loss_rpn_box_reg: 0.0085 (0.0218)  time: 0.6054  data: 0.0078  max mem: 5190
Epoch: [9]  [39/40]  eta: 0:00:00  lr: 0.005000  loss: 0.2970 (0.2839)  loss_classifier: 0.0435 (0.0397)  loss_box_reg: 0.0993 (0.0879)  loss_mask: 0.1364 (0.1355)  loss_objectness: 0.0016 (0.0021)  loss_rpn_box_reg: 0.0075 (0.0188)  time: 0.6129  data: 0.0078  max mem: 5190
Epoch: [9] Total time: 0:00:24 (0.6138 s / it)
creating index...
index created!
Test:  [ 0/10]  eta: 0:00:07  model_time: 0.2418 (0.2418)  evaluator_time: 0.0310 (0.0310)  time: 0.7868  data: 0.5110  max mem: 5190
Test:  [ 9/10]  eta: 0:00:00  model_time: 0.2108 (0.2142)  evaluator_time: 0.0237 (0.0237)  time: 0.2964  data: 0.0557  max mem: 5190
Test: Total time: 0:00:03 (0.3025 s / it)
Averaged stats: model_time: 0.2108 (0.2142)  evaluator_time: 0.0237 (0.0237)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.509
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.846
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.527
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.305
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.512
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.680
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.235
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.590
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.590
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.432
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.588
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.730
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.462
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.815
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.456
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.192
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.464
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.637
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.219
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.547
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.547
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.437
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.540
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.652
Epoch: [10]  [ 0/40]  eta: 0:00:47  lr: 0.000500  loss: 0.3321 (0.3321)  loss_classifier: 0.0440 (0.0440)  loss_box_reg: 0.1080 (0.1080)  loss_mask: 0.1587 (0.1587)  loss_objectness: 0.0030 (0.0030)  loss_rpn_box_reg: 0.0183 (0.0183)  time: 1.1958  data: 0.7041  max mem: 5190
Epoch: [10]  [10/40]  eta: 0:00:18  lr: 0.000500  loss: 0.2306 (0.2487)  loss_classifier: 0.0328 (0.0343)  loss_box_reg: 0.0713 (0.0790)  loss_mask: 0.1195 (0.1283)  loss_objectness: 0.0017 (0.0018)  loss_rpn_box_reg: 0.0040 (0.0053)  time: 0.6297  data: 0.0708  max mem: 5190
Epoch: [10]  [20/40]  eta: 0:00:12  lr: 0.000500  loss: 0.2306 (0.2417)  loss_classifier: 0.0318 (0.0344)  loss_box_reg: 0.0674 (0.0744)  loss_mask: 0.1125 (0.1233)  loss_objectness: 0.0013 (0.0019)  loss_rpn_box_reg: 0.0044 (0.0077)  time: 0.5867  data: 0.0075  max mem: 5190
Epoch: [10]  [30/40]  eta: 0:00:06  lr: 0.000500  loss: 0.2412 (0.2437)  loss_classifier: 0.0337 (0.0351)  loss_box_reg: 0.0692 (0.0751)  loss_mask: 0.1215 (0.1248)  loss_objectness: 0.0013 (0.0017)  loss_rpn_box_reg: 0.0065 (0.0071)  time: 0.6076  data: 0.0075  max mem: 5190
Epoch: [10]  [39/40]  eta: 0:00:00  lr: 0.000500  loss: 0.2475 (0.2550)  loss_classifier: 0.0365 (0.0367)  loss_box_reg: 0.0773 (0.0779)  loss_mask: 0.1252 (0.1270)  loss_objectness: 0.0011 (0.0017)  loss_rpn_box_reg: 0.0053 (0.0116)  time: 0.6154  data: 0.0075  max mem: 5190
Epoch: [10] Total time: 0:00:24 (0.6175 s / it)
creating index...
index created!
Test:  [ 0/10]  eta: 0:00:07  model_time: 0.2341 (0.2341)  evaluator_time: 0.0288 (0.0288)  time: 0.7587  data: 0.4926  max mem: 5190
Test:  [ 9/10]  eta: 0:00:00  model_time: 0.2070 (0.2116)  evaluator_time: 0.0228 (0.0232)  time: 0.2912  data: 0.0535  max mem: 5190
Test: Total time: 0:00:02 (0.2968 s / it)
Averaged stats: model_time: 0.2070 (0.2116)  evaluator_time: 0.0228 (0.0232)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.508
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.844
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.523
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.314
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.514
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.668
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.229
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.597
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.597
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.484
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.587
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.739
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.455
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.839
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.444
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.182
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.464
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.618
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.215
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.547
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.549
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.458
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.533
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.665
Epoch: [11]  [ 0/40]  eta: 0:00:46  lr: 0.000500  loss: 0.2035 (0.2035)  loss_classifier: 0.0314 (0.0314)  loss_box_reg: 0.0525 (0.0525)  loss_mask: 0.1153 (0.1153)  loss_objectness: 0.0004 (0.0004)  loss_rpn_box_reg: 0.0038 (0.0038)  time: 1.1690  data: 0.6649  max mem: 5190
Epoch: [11]  [10/40]  eta: 0:00:18  lr: 0.000500  loss: 0.2174 (0.2344)  loss_classifier: 0.0302 (0.0337)  loss_box_reg: 0.0555 (0.0690)  loss_mask: 0.1240 (0.1247)  loss_objectness: 0.0009 (0.0011)  loss_rpn_box_reg: 0.0038 (0.0059)  time: 0.6311  data: 0.0677  max mem: 5190
Epoch: [11]  [20/40]  eta: 0:00:12  lr: 0.000500  loss: 0.2174 (0.2293)  loss_classifier: 0.0291 (0.0326)  loss_box_reg: 0.0588 (0.0660)  loss_mask: 0.1240 (0.1237)  loss_objectness: 0.0008 (0.0011)  loss_rpn_box_reg: 0.0038 (0.0059)  time: 0.5906  data: 0.0080  max mem: 5190
Epoch: [11]  [30/40]  eta: 0:00:06  lr: 0.000500  loss: 0.2337 (0.2476)  loss_classifier: 0.0312 (0.0350)  loss_box_reg: 0.0716 (0.0697)  loss_mask: 0.1262 (0.1259)  loss_objectness: 0.0008 (0.0019)  loss_rpn_box_reg: 0.0041 (0.0151)  time: 0.6069  data: 0.0079  max mem: 5190
Epoch: [11]  [39/40]  eta: 0:00:00  lr: 0.000500  loss: 0.2337 (0.2495)  loss_classifier: 0.0383 (0.0353)  loss_box_reg: 0.0788 (0.0721)  loss_mask: 0.1262 (0.1265)  loss_objectness: 0.0013 (0.0022)  loss_rpn_box_reg: 0.0043 (0.0135)  time: 0.6157  data: 0.0077  max mem: 5190
Epoch: [11] Total time: 0:00:24 (0.6188 s / it)
creating index...
index created!
Test:  [ 0/10]  eta: 0:00:07  model_time: 0.2466 (0.2466)  evaluator_time: 0.0265 (0.0265)  time: 0.7916  data: 0.5153  max mem: 5190
Test:  [ 9/10]  eta: 0:00:00  model_time: 0.2073 (0.2125)  evaluator_time: 0.0216 (0.0228)  time: 0.2950  data: 0.0568  max mem: 5190
Test: Total time: 0:00:03 (0.3009 s / it)
Averaged stats: model_time: 0.2073 (0.2125)  evaluator_time: 0.0216 (0.0228)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.506
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.846
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.527
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.299
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.518
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.638
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.234
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.586
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.586
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.447
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.589
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.700
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.457
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.828
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.444
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.182
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.465
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.625
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.214
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.553
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.553
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.463
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.539
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.665
Epoch: [12]  [ 0/40]  eta: 0:00:51  lr: 0.000500  loss: 0.3570 (0.3570)  loss_classifier: 0.0595 (0.0595)  loss_box_reg: 0.1149 (0.1149)  loss_mask: 0.1768 (0.1768)  loss_objectness: 0.0016 (0.0016)  loss_rpn_box_reg: 0.0042 (0.0042)  time: 1.2779  data: 0.7711  max mem: 5190
Epoch: [12]  [10/40]  eta: 0:00:19  lr: 0.000500  loss: 0.2179 (0.2348)  loss_classifier: 0.0281 (0.0352)  loss_box_reg: 0.0571 (0.0694)  loss_mask: 0.1146 (0.1191)  loss_objectness: 0.0016 (0.0021)  loss_rpn_box_reg: 0.0075 (0.0090)  time: 0.6373  data: 0.0754  max mem: 5190
Epoch: [12]  [20/40]  eta: 0:00:12  lr: 0.000500  loss: 0.2297 (0.2334)  loss_classifier: 0.0281 (0.0334)  loss_box_reg: 0.0571 (0.0658)  loss_mask: 0.1186 (0.1254)  loss_objectness: 0.0012 (0.0017)  loss_rpn_box_reg: 0.0053 (0.0069)  time: 0.5885  data: 0.0068  max mem: 5190
Epoch: [12]  [30/40]  eta: 0:00:06  lr: 0.000500  loss: 0.2497 (0.2481)  loss_classifier: 0.0351 (0.0349)  loss_box_reg: 0.0658 (0.0693)  loss_mask: 0.1264 (0.1265)  loss_objectness: 0.0011 (0.0020)  loss_rpn_box_reg: 0.0032 (0.0153)  time: 0.6101  data: 0.0078  max mem: 5190
Epoch: [12]  [39/40]  eta: 0:00:00  lr: 0.000500  loss: 0.2469 (0.2523)  loss_classifier: 0.0351 (0.0351)  loss_box_reg: 0.0689 (0.0732)  loss_mask: 0.1258 (0.1284)  loss_objectness: 0.0015 (0.0021)  loss_rpn_box_reg: 0.0035 (0.0135)  time: 0.6176  data: 0.0077  max mem: 5190
Epoch: [12] Total time: 0:00:24 (0.6217 s / it)
creating index...
index created!
Test:  [ 0/10]  eta: 0:00:07  model_time: 0.2384 (0.2384)  evaluator_time: 0.0313 (0.0313)  time: 0.7826  data: 0.5098  max mem: 5190
Test:  [ 9/10]  eta: 0:00:00  model_time: 0.2071 (0.2114)  evaluator_time: 0.0212 (0.0228)  time: 0.2926  data: 0.0556  max mem: 5190
Test: Total time: 0:00:02 (0.2985 s / it)
Averaged stats: model_time: 0.2071 (0.2114)  evaluator_time: 0.0212 (0.0228)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.511
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.839
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.523
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.298
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.526
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.640
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.245
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.590
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.590
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.442
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.595
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.704
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.460
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.828
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.450
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.187
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.474
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.603
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.212
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.550
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.550
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.458
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.540
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.648
Epoch: [13]  [ 0/40]  eta: 0:00:46  lr: 0.000500  loss: 0.2275 (0.2275)  loss_classifier: 0.0289 (0.0289)  loss_box_reg: 0.0634 (0.0634)  loss_mask: 0.1276 (0.1276)  loss_objectness: 0.0006 (0.0006)  loss_rpn_box_reg: 0.0071 (0.0071)  time: 1.1639  data: 0.6723  max mem: 5190
Epoch: [13]  [10/40]  eta: 0:00:19  lr: 0.000500  loss: 0.2275 (0.2309)  loss_classifier: 0.0289 (0.0288)  loss_box_reg: 0.0691 (0.0649)  loss_mask: 0.1231 (0.1292)  loss_objectness: 0.0014 (0.0024)  loss_rpn_box_reg: 0.0051 (0.0056)  time: 0.6364  data: 0.0684  max mem: 5190
Epoch: [13]  [20/40]  eta: 0:00:12  lr: 0.000500  loss: 0.2297 (0.2264)  loss_classifier: 0.0304 (0.0313)  loss_box_reg: 0.0692 (0.0662)  loss_mask: 0.1195 (0.1214)  loss_objectness: 0.0009 (0.0021)  loss_rpn_box_reg: 0.0042 (0.0054)  time: 0.5977  data: 0.0079  max mem: 5190
Epoch: [13]  [30/40]  eta: 0:00:06  lr: 0.000500  loss: 0.2516 (0.2475)  loss_classifier: 0.0347 (0.0338)  loss_box_reg: 0.0748 (0.0712)  loss_mask: 0.1243 (0.1254)  loss_objectness: 0.0014 (0.0024)  loss_rpn_box_reg: 0.0042 (0.0147)  time: 0.6137  data: 0.0078  max mem: 5190
Epoch: [13]  [39/40]  eta: 0:00:00  lr: 0.000500  loss: 0.2674 (0.2471)  loss_classifier: 0.0375 (0.0338)  loss_box_reg: 0.0769 (0.0716)  loss_mask: 0.1261 (0.1266)  loss_objectness: 0.0019 (0.0022)  loss_rpn_box_reg: 0.0061 (0.0129)  time: 0.6152  data: 0.0077  max mem: 5190
Epoch: [13] Total time: 0:00:24 (0.6218 s / it)
creating index...
index created!
Test:  [ 0/10]  eta: 0:00:07  model_time: 0.2262 (0.2262)  evaluator_time: 0.0280 (0.0280)  time: 0.7557  data: 0.4983  max mem: 5190
Test:  [ 9/10]  eta: 0:00:00  model_time: 0.2076 (0.2100)  evaluator_time: 0.0208 (0.0223)  time: 0.2903  data: 0.0550  max mem: 5190
Test: Total time: 0:00:02 (0.2967 s / it)
Averaged stats: model_time: 0.2076 (0.2100)  evaluator_time: 0.0208 (0.0223)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.503
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.848
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.530
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.292
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.522
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.636
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.239
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.581
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.581
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.416
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.592
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.700
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.471
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.842
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.469
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.181
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.488
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.624
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.217
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.558
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.558
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.432
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.551
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.674
Epoch: [14]  [ 0/40]  eta: 0:00:47  lr: 0.000500  loss: 0.1741 (0.1741)  loss_classifier: 0.0218 (0.0218)  loss_box_reg: 0.0544 (0.0544)  loss_mask: 0.0942 (0.0942)  loss_objectness: 0.0019 (0.0019)  loss_rpn_box_reg: 0.0018 (0.0018)  time: 1.1930  data: 0.6933  max mem: 5190
Epoch: [14]  [10/40]  eta: 0:00:19  lr: 0.000500  loss: 0.2146 (0.2241)  loss_classifier: 0.0331 (0.0341)  loss_box_reg: 0.0594 (0.0659)  loss_mask: 0.1116 (0.1187)  loss_objectness: 0.0015 (0.0014)  loss_rpn_box_reg: 0.0023 (0.0040)  time: 0.6368  data: 0.0699  max mem: 5190
Epoch: [14]  [20/40]  eta: 0:00:12  lr: 0.000500  loss: 0.2088 (0.2202)  loss_classifier: 0.0298 (0.0316)  loss_box_reg: 0.0577 (0.0618)  loss_mask: 0.1152 (0.1212)  loss_objectness: 0.0006 (0.0013)  loss_rpn_box_reg: 0.0025 (0.0043)  time: 0.5918  data: 0.0075  max mem: 5190
Epoch: [14]  [30/40]  eta: 0:00:06  lr: 0.000500  loss: 0.2371 (0.2380)  loss_classifier: 0.0306 (0.0334)  loss_box_reg: 0.0566 (0.0666)  loss_mask: 0.1207 (0.1230)  loss_objectness: 0.0009 (0.0018)  loss_rpn_box_reg: 0.0047 (0.0133)  time: 0.6113  data: 0.0074  max mem: 5190
Epoch: [14]  [39/40]  eta: 0:00:00  lr: 0.000500  loss: 0.2501 (0.2438)  loss_classifier: 0.0373 (0.0353)  loss_box_reg: 0.0720 (0.0685)  loss_mask: 0.1260 (0.1259)  loss_objectness: 0.0011 (0.0019)  loss_rpn_box_reg: 0.0070 (0.0123)  time: 0.6194  data: 0.0072  max mem: 5190
Epoch: [14] Total time: 0:00:24 (0.6222 s / it)
creating index...
index created!
Test:  [ 0/10]  eta: 0:00:07  model_time: 0.2269 (0.2269)  evaluator_time: 0.0232 (0.0232)  time: 0.7710  data: 0.5176  max mem: 5190
Test:  [ 9/10]  eta: 0:00:00  model_time: 0.2077 (0.2095)  evaluator_time: 0.0206 (0.0210)  time: 0.2900  data: 0.0566  max mem: 5190
Test: Total time: 0:00:02 (0.2958 s / it)
Averaged stats: model_time: 0.2077 (0.2095)  evaluator_time: 0.0206 (0.0210)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.495
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.839
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.506
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.292
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.513
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.619
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.243
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.572
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.572
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.416
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.579
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.687
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.453
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.829
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.441
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.174
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.468
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.620
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.213
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.544
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.544
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.432
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.529
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.674
Epoch: [15]  [ 0/40]  eta: 0:00:48  lr: 0.000500  loss: 0.2820 (0.2820)  loss_classifier: 0.0364 (0.0364)  loss_box_reg: 0.0847 (0.0847)  loss_mask: 0.1516 (0.1516)  loss_objectness: 0.0037 (0.0037)  loss_rpn_box_reg: 0.0055 (0.0055)  time: 1.2072  data: 0.7219  max mem: 5190
Epoch: [15]  [10/40]  eta: 0:00:19  lr: 0.000500  loss: 0.2736 (0.2588)  loss_classifier: 0.0356 (0.0388)  loss_box_reg: 0.0792 (0.0782)  loss_mask: 0.1396 (0.1347)  loss_objectness: 0.0016 (0.0020)  loss_rpn_box_reg: 0.0034 (0.0051)  time: 0.6397  data: 0.0724  max mem: 5190
Epoch: [15]  [20/40]  eta: 0:00:12  lr: 0.000500  loss: 0.2405 (0.2455)  loss_classifier: 0.0326 (0.0358)  loss_box_reg: 0.0674 (0.0752)  loss_mask: 0.1295 (0.1283)  loss_objectness: 0.0005 (0.0013)  loss_rpn_box_reg: 0.0034 (0.0049)  time: 0.5945  data: 0.0073  max mem: 5190
Epoch: [15]  [30/40]  eta: 0:00:06  lr: 0.000500  loss: 0.2194 (0.2326)  loss_classifier: 0.0254 (0.0329)  loss_box_reg: 0.0592 (0.0688)  loss_mask: 0.1149 (0.1242)  loss_objectness: 0.0008 (0.0015)  loss_rpn_box_reg: 0.0043 (0.0053)  time: 0.6093  data: 0.0073  max mem: 5190
Epoch: [15]  [39/40]  eta: 0:00:00  lr: 0.000500  loss: 0.2208 (0.2349)  loss_classifier: 0.0311 (0.0329)  loss_box_reg: 0.0620 (0.0678)  loss_mask: 0.1149 (0.1229)  loss_objectness: 0.0008 (0.0016)  loss_rpn_box_reg: 0.0054 (0.0097)  time: 0.6177  data: 0.0075  max mem: 5190
Epoch: [15] Total time: 0:00:24 (0.6226 s / it)
creating index...
index created!
Test:  [ 0/10]  eta: 0:00:07  model_time: 0.2261 (0.2261)  evaluator_time: 0.0241 (0.0241)  time: 0.7789  data: 0.5254  max mem: 5190
Test:  [ 9/10]  eta: 0:00:00  model_time: 0.2057 (0.2090)  evaluator_time: 0.0199 (0.0207)  time: 0.2904  data: 0.0577  max mem: 5190
Test: Total time: 0:00:02 (0.2961 s / it)
Averaged stats: model_time: 0.2057 (0.2090)  evaluator_time: 0.0199 (0.0207)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.507
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.846
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.525
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.295
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.530
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.620
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.245
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.578
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.580
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.405
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.597
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.683
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.462
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.837
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.467
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.181
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.480
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.615
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.210
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.547
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.547
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.416
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.543
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.657
Epoch: [16]  [ 0/40]  eta: 0:00:47  lr: 0.000500  loss: 0.3187 (0.3187)  loss_classifier: 0.0470 (0.0470)  loss_box_reg: 0.0923 (0.0923)  loss_mask: 0.1577 (0.1577)  loss_objectness: 0.0040 (0.0040)  loss_rpn_box_reg: 0.0177 (0.0177)  time: 1.1972  data: 0.6992  max mem: 5190
Epoch: [16]  [10/40]  eta: 0:00:19  lr: 0.000500  loss: 0.2466 (0.2551)  loss_classifier: 0.0432 (0.0379)  loss_box_reg: 0.0893 (0.0785)  loss_mask: 0.1319 (0.1307)  loss_objectness: 0.0012 (0.0014)  loss_rpn_box_reg: 0.0054 (0.0066)  time: 0.6415  data: 0.0705  max mem: 5190
Epoch: [16]  [20/40]  eta: 0:00:12  lr: 0.000500  loss: 0.2057 (0.2344)  loss_classifier: 0.0285 (0.0338)  loss_box_reg: 0.0562 (0.0657)  loss_mask: 0.1138 (0.1213)  loss_objectness: 0.0012 (0.0016)  loss_rpn_box_reg: 0.0048 (0.0121)  time: 0.5997  data: 0.0077  max mem: 5190
Epoch: [16]  [30/40]  eta: 0:00:06  lr: 0.000500  loss: 0.1889 (0.2253)  loss_classifier: 0.0277 (0.0319)  loss_box_reg: 0.0488 (0.0626)  loss_mask: 0.1138 (0.1199)  loss_objectness: 0.0010 (0.0014)  loss_rpn_box_reg: 0.0030 (0.0095)  time: 0.6131  data: 0.0078  max mem: 5190
Epoch: [16]  [39/40]  eta: 0:00:00  lr: 0.000500  loss: 0.2187 (0.2350)  loss_classifier: 0.0299 (0.0335)  loss_box_reg: 0.0605 (0.0671)  loss_mask: 0.1178 (0.1235)  loss_objectness: 0.0010 (0.0016)  loss_rpn_box_reg: 0.0035 (0.0093)  time: 0.6179  data: 0.0078  max mem: 5190
Epoch: [16] Total time: 0:00:25 (0.6252 s / it)
creating index...
index created!
Test:  [ 0/10]  eta: 0:00:07  model_time: 0.2340 (0.2340)  evaluator_time: 0.0250 (0.0250)  time: 0.7698  data: 0.5075  max mem: 5190
Test:  [ 9/10]  eta: 0:00:00  model_time: 0.2057 (0.2088)  evaluator_time: 0.0195 (0.0201)  time: 0.2875  data: 0.0556  max mem: 5190
Test: Total time: 0:00:02 (0.2937 s / it)
Averaged stats: model_time: 0.2057 (0.2088)  evaluator_time: 0.0195 (0.0201)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.501
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.849
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.505
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.309
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.519
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.632
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.240
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.580
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.580
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.421
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.587
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.700
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.458
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.827
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.465
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.202
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.473
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.603
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.209
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.542
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.542
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.453
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.531
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.643
Epoch: [17]  [ 0/40]  eta: 0:00:47  lr: 0.000500  loss: 0.2313 (0.2313)  loss_classifier: 0.0295 (0.0295)  loss_box_reg: 0.0676 (0.0676)  loss_mask: 0.1281 (0.1281)  loss_objectness: 0.0005 (0.0005)  loss_rpn_box_reg: 0.0057 (0.0057)  time: 1.1759  data: 0.6829  max mem: 5190
Epoch: [17]  [10/40]  eta: 0:00:19  lr: 0.000500  loss: 0.2311 (0.2677)  loss_classifier: 0.0329 (0.0356)  loss_box_reg: 0.0644 (0.0722)  loss_mask: 0.1313 (0.1307)  loss_objectness: 0.0014 (0.0025)  loss_rpn_box_reg: 0.0032 (0.0269)  time: 0.6392  data: 0.0721  max mem: 5190
Epoch: [17]  [20/40]  eta: 0:00:12  lr: 0.000500  loss: 0.2159 (0.2499)  loss_classifier: 0.0317 (0.0340)  loss_box_reg: 0.0625 (0.0704)  loss_mask: 0.1313 (0.1264)  loss_objectness: 0.0008 (0.0019)  loss_rpn_box_reg: 0.0032 (0.0171)  time: 0.5984  data: 0.0092  max mem: 5190
Epoch: [17]  [30/40]  eta: 0:00:06  lr: 0.000500  loss: 0.2114 (0.2363)  loss_classifier: 0.0304 (0.0333)  loss_box_reg: 0.0577 (0.0664)  loss_mask: 0.1172 (0.1223)  loss_objectness: 0.0006 (0.0016)  loss_rpn_box_reg: 0.0037 (0.0127)  time: 0.6168  data: 0.0076  max mem: 5190
Epoch: [17]  [39/40]  eta: 0:00:00  lr: 0.000500  loss: 0.1931 (0.2321)  loss_classifier: 0.0316 (0.0327)  loss_box_reg: 0.0566 (0.0649)  loss_mask: 0.1115 (0.1215)  loss_objectness: 0.0006 (0.0016)  loss_rpn_box_reg: 0.0037 (0.0114)  time: 0.6197  data: 0.0078  max mem: 5190
Epoch: [17] Total time: 0:00:24 (0.6247 s / it)
creating index...
index created!
Test:  [ 0/10]  eta: 0:00:07  model_time: 0.2365 (0.2365)  evaluator_time: 0.0254 (0.0254)  time: 0.7710  data: 0.5060  max mem: 5190
Test:  [ 9/10]  eta: 0:00:00  model_time: 0.2052 (0.2091)  evaluator_time: 0.0190 (0.0194)  time: 0.2871  data: 0.0557  max mem: 5190
Test: Total time: 0:00:02 (0.2938 s / it)
Averaged stats: model_time: 0.2052 (0.2091)  evaluator_time: 0.0190 (0.0194)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.503
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.843
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.510
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.302
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.519
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.638
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.228
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.581
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.581
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.421
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.589
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.700
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.459
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.832
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.456
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.189
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.478
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.600
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.212
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.542
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.542
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.432
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.536
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.643
Epoch: [18]  [ 0/40]  eta: 0:00:49  lr: 0.000500  loss: 0.1389 (0.1389)  loss_classifier: 0.0246 (0.0246)  loss_box_reg: 0.0311 (0.0311)  loss_mask: 0.0781 (0.0781)  loss_objectness: 0.0010 (0.0010)  loss_rpn_box_reg: 0.0040 (0.0040)  time: 1.2488  data: 0.7533  max mem: 5190
Epoch: [18]  [10/40]  eta: 0:00:19  lr: 0.000500  loss: 0.2122 (0.2432)  loss_classifier: 0.0293 (0.0333)  loss_box_reg: 0.0531 (0.0733)  loss_mask: 0.1174 (0.1266)  loss_objectness: 0.0021 (0.0023)  loss_rpn_box_reg: 0.0070 (0.0077)  time: 0.6452  data: 0.0741  max mem: 5190
Epoch: [18]  [20/40]  eta: 0:00:12  lr: 0.000500  loss: 0.2122 (0.2323)  loss_classifier: 0.0280 (0.0325)  loss_box_reg: 0.0610 (0.0694)  loss_mask: 0.1146 (0.1224)  loss_objectness: 0.0010 (0.0017)  loss_rpn_box_reg: 0.0029 (0.0062)  time: 0.5998  data: 0.0067  max mem: 5190
Epoch: [18]  [30/40]  eta: 0:00:06  lr: 0.000500  loss: 0.2194 (0.2341)  loss_classifier: 0.0324 (0.0323)  loss_box_reg: 0.0615 (0.0663)  loss_mask: 0.1136 (0.1208)  loss_objectness: 0.0006 (0.0016)  loss_rpn_box_reg: 0.0028 (0.0132)  time: 0.6141  data: 0.0074  max mem: 5190
Epoch: [18]  [39/40]  eta: 0:00:00  lr: 0.000500  loss: 0.2194 (0.2331)  loss_classifier: 0.0324 (0.0326)  loss_box_reg: 0.0604 (0.0653)  loss_mask: 0.1180 (0.1226)  loss_objectness: 0.0007 (0.0015)  loss_rpn_box_reg: 0.0041 (0.0112)  time: 0.6128  data: 0.0076  max mem: 5190
Epoch: [18] Total time: 0:00:24 (0.6241 s / it)
creating index...
index created!
Test:  [ 0/10]  eta: 0:00:07  model_time: 0.2258 (0.2258)  evaluator_time: 0.0291 (0.0291)  time: 0.7811  data: 0.5229  max mem: 5190
Test:  [ 9/10]  eta: 0:00:00  model_time: 0.2061 (0.2082)  evaluator_time: 0.0199 (0.0207)  time: 0.2885  data: 0.0567  max mem: 5190
Test: Total time: 0:00:02 (0.2941 s / it)
Averaged stats: model_time: 0.2061 (0.2082)  evaluator_time: 0.0199 (0.0207)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.512
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.840
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.526
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.310
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.523
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.669
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.239
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.584
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.584
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.416
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.588
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.730
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.462
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.834
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.470
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.198
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.469
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.608
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.215
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.546
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.546
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.421
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.541
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.652
Epoch: [19]  [ 0/40]  eta: 0:00:50  lr: 0.000500  loss: 0.1895 (0.1895)  loss_classifier: 0.0305 (0.0305)  loss_box_reg: 0.0564 (0.0564)  loss_mask: 0.0994 (0.0994)  loss_objectness: 0.0007 (0.0007)  loss_rpn_box_reg: 0.0025 (0.0025)  time: 1.2744  data: 0.7670  max mem: 5190
Epoch: [19]  [10/40]  eta: 0:00:19  lr: 0.000500  loss: 0.1949 (0.2342)  loss_classifier: 0.0300 (0.0327)  loss_box_reg: 0.0547 (0.0607)  loss_mask: 0.1105 (0.1160)  loss_objectness: 0.0007 (0.0015)  loss_rpn_box_reg: 0.0025 (0.0233)  time: 0.6384  data: 0.0755  max mem: 5190
Epoch: [19]  [20/40]  eta: 0:00:12  lr: 0.000500  loss: 0.2128 (0.2312)  loss_classifier: 0.0313 (0.0329)  loss_box_reg: 0.0546 (0.0607)  loss_mask: 0.1210 (0.1211)  loss_objectness: 0.0010 (0.0014)  loss_rpn_box_reg: 0.0038 (0.0150)  time: 0.5912  data: 0.0070  max mem: 5190
Epoch: [19]  [30/40]  eta: 0:00:06  lr: 0.000500  loss: 0.2133 (0.2261)  loss_classifier: 0.0282 (0.0313)  loss_box_reg: 0.0546 (0.0613)  loss_mask: 0.1152 (0.1204)  loss_objectness: 0.0011 (0.0013)  loss_rpn_box_reg: 0.0038 (0.0119)  time: 0.6125  data: 0.0074  max mem: 5190
Epoch: [19]  [39/40]  eta: 0:00:00  lr: 0.000500  loss: 0.2221 (0.2296)  loss_classifier: 0.0327 (0.0329)  loss_box_reg: 0.0637 (0.0645)  loss_mask: 0.1152 (0.1204)  loss_objectness: 0.0008 (0.0012)  loss_rpn_box_reg: 0.0044 (0.0107)  time: 0.6198  data: 0.0074  max mem: 5190
Epoch: [19] Total time: 0:00:24 (0.6238 s / it)
creating index...
index created!
Test:  [ 0/10]  eta: 0:00:07  model_time: 0.2336 (0.2336)  evaluator_time: 0.0330 (0.0330)  time: 0.7830  data: 0.5135  max mem: 5190
Test:  [ 9/10]  eta: 0:00:00  model_time: 0.2039 (0.2076)  evaluator_time: 0.0177 (0.0197)  time: 0.2866  data: 0.0564  max mem: 5190
Test: Total time: 0:00:02 (0.2935 s / it)
Averaged stats: model_time: 0.2039 (0.2076)  evaluator_time: 0.0177 (0.0197)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.499
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.845
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.505
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.300
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.521
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.620
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.234
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.574
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.574
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.405
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.591
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.674
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.450
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.811
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.463
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.188
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.466
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.593
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.206
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.541
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.541
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.432
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.535
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.643
{% endraw %} {% raw %} 
# pick one image from the test set
img, _ = data_loader_test.dataset[0]
# put the model in evaluation mode
model.eval()
with torch.no_grad():
    prediction = model([img.to(device)])
prediction

[{'boxes': tensor([[296.2079, 193.3952, 453.0812, 267.1093],
          [442.2224, 187.0209, 616.6365, 252.8787],
          [295.5228, 212.7988, 341.5661, 258.0097],
          [295.8234, 197.4667, 370.7387, 262.2480],
          [369.6344, 216.4189, 417.1077, 230.0708],
          [326.0764, 192.2509, 429.9577, 242.1528]], device='cuda:0'),
  'labels': tensor([1, 1, 1, 1, 1, 1], device='cuda:0'),
  'scores': tensor([0.9996, 0.9995, 0.9818, 0.2625, 0.2471, 0.2113], device='cuda:0'),
  'masks': tensor([[[[0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.],
            ...,
            [0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.]]],
  
  
          [[[0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.],
            ...,
            [0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.]]],
  
  
          [[[0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.],
            ...,
            [0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.]]],
  
  
          [[[0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.],
            ...,
            [0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.]]],
  
  
          [[[0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.],
            ...,
            [0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.]]],
  
  
          [[[0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.],
            ...,
            [0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.]]]], device='cuda:0')}]
{% endraw %} {% raw %} 
from dolphins_recognition_challenge.datasets import stack_imgs

def show_pred(dl, n=None, score_limit=0.5, width=600):
    dataset_test = dl.dataset
    if n == None:
        n = len(dataset_test)

    for i in range(n):
        img = dataset_test[i][0]
        img_bg = Image.fromarray(img.mul(255).permute(1, 2, 0).byte().numpy())
        images = [img_bg]
        model.eval()
        with torch.no_grad():
            prediction = model([img.to(device)])
        predicted_masks = prediction[0]["masks"]
        scores = prediction[0]["scores"]

        for i in range(predicted_masks.shape[0]):
            score = scores[i]
            if score >= score_limit:
                bg = img_bg.copy()
                fg = Image.fromarray(predicted_masks[i, 0].mul(255).byte().cpu().numpy())
                bg.paste(fg.convert("RGB"), (0, 0), fg)
                images.append(bg)
        
        display(stack_imgs(images, width))


show_pred(data_loader_test, score_limit=0.5, width=1200)
{% endraw %} {% raw %} 
def _save_model_with_timestamp(
    model, save_path="/work/data/dupini/processed/body_100_resized/"
):
    save_date_path = (
        save_path + "model" + datetime.now().strftime("-%Y-%m-%d-%H-%M-%S") + ".pt"
    )
    print(save_date_path)
    torch.save(model.state_dict(), save_date_path)
{% endraw %} {% raw %}
{% endraw %}

Calculate metrics